EPA Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmissions and Storage Source Categories, July 2011

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE

1

Draft Residual Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories

by EPAs Office of Air Quality Planning and Standards Office of Air and Radiation July 2011


2

Table of Contents1 2 Introduction ........................................................................................................................ 4 Methods .............................................................................................................................. 5 2.1 Emissions and source data ......................................................................................... 5 2.2 Dispersion modeling for inhalation exposure assessment ......................................... 5 2.3 Estimating human inhalation exposure ...................................................................... 8 2.4 Acute Risk Screening and Refined Assessments ....................................................... 8 2.5 Multipathway and environmental risk screening ..................................................... 10 2.6 Dose-Response Assessment ..................................................................................... 11 2.6.1 Sources of chronic dose-response information ................................................ 11 2.6.2 Sources of acute dose-response information .................................................... 17 2.7 Risk Characterization ............................................................................................... 21 2.7.1 General ............................................................................................................. 21 2.7.2 Mixtures ........................................................................................................... 23 2.7.3 Facility-wide Risks .......................................................................................... 23 3 Risk Results for the Natural Gas Transmission and Storage Source Category........... 24 3.1 Source Category Description and Results ................................................................ 24 3.2 Risk Characterization ............................................................................................... 26 4 Risk Results for the Oil and Natural Gas Production Source Category....................... 30 4.1 Source Category Description and Results ................................................................ 30 4.2 Risk Characterization ............................................................................................... 32 5 General Discussion of Uncertainties and How They Have Been Addressed................... 36 5.1 Exposure Modeling Uncertainties ............................................................................ 36 5.2 Uncertainties in the Dose-Response Relationships .................................................. 37 5 References ........................................................................................................................ 45 Index of Tables Table 2.2-1 AERMOD version 09292 model options for RTR modeling ................................ 6 Table 2.6-1 (a) Dose-Response Values for Chronic Inhalation Exposure to Carcinogens .... 14 Table 2.6-1 (b) Dose-Response Values for Chronic Inhalation Exposure to Noncarcinogens .................................................................................................................................................. 15 Table 2.6-2 Dose-Response Values for Acute Exposure ........................................................ 20 Table 3.1-1 Summary of Emissions from the Natural Gas Transmission and Storage Source Category and Availability of Dose-Response Values .............................................................. 25 Table 3.2-1 Summary of Source Category Level Inhalation Risks for Natural Gas Transmission and Storage ........................................................................................................ 28 Table 3.2-2 Summary of Refined Acute Results for Natural Gas Transmission and Storage Facilities ................................................................................................................................... 29 Table 3.2-3 Source Category Contribution to Facility-Wide Cancer Risks ........................... 29 Table 4.1-1 Summary of Emissions from the Oil and Natural Gas Production Source Category and Availability of Dose-Response Values .............................................................. 31 Table 4.2-1 Summary of Source Category Level Inhalation Risks for Oil and Natural Gas Production ................................................................................................................................ 34 Table 4.2-2 Summary of Refined Acute Results for Oil and Natural Gas Production ........... 34

3 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 4.2-3 Source Category Contribution to Facility-Wide Cancer Risks ........................... 36 Appendices Appendix 1 Appendix 2 Appendix 3 Appendix 4 Appendix 5 Appendix 6 Appendix 7 Emissions Inventory Support Memorandum Technical Support Document for HEM-AERMOD Modeling Meteorological Data for HEM-AERMOD Modeling Analysis of data on short-term emission rates relative to long-term emission rates Technical Support Document for TRIM-Based Multipathway Screening Scenario for RTR: Summary of Approach and Evaluation Detailed Risk Modeling Results Acute Impacts Refined Analysis Figures

Index of Acronyms AERMOD AEGL ASTDR CalEPA ERPG HAP HEM HI HQ IRIS MACT MIR MOA NAC NATA NEI NPRM PB-HAP POM REL RfC RfD RTR TOSHI URE American Meteorological Society/EPA Regulatory Model Acute exposure guideline level US Agency for Toxic Substances and Disease Registry California Environmental Agency Emergency Response Planning Guideline Hazardous Air Pollutant Human Exposure Model Hazard index Hazard quotient Integrated Risk Information System Maximum Achievable Control Technology Maximum Individual Risk Mode of action National Advisory Committee National Air Toxics Assessment National Emissions Inventory Notice of Proposed Rulemaking Persistent and Bioaccumulative - HAP Polycyclic organic matter Reference exposure level Reference concentration Reference dose Risk and Technology Target-organ-specific hazard index Unit risk estimate


4

1 IntroductionSection 112 of the Clean Air Act (CAA) establishes a two-stage regulatory process for addressing emissions of hazardous air pollutants (HAPs) from stationary sources. In the first stage, section 112(d) requires the Environmental Protection Agency (EPA, or the Agency) to develop technology-based standards for categories of sources (e.g., petroleum refineries, pulp and paper mills, etc.) [1]. EPA has largely completed the initial Maximum Achievable Control Technology (MACT) standards as required under this provision. Under section 112(d)(6), EPA must review each of these technology-based standards at least every eight years and revise a standard, as necessary, taking into account developments in practices, processes and control technologies. In the second stage, EPA is required under section 112(f)(2) to assess the health and environmental risks that remain after implementation of the MACT standards. If additional risk reductions are necessary to protect public health with an ample margin of safety or to prevent an adverse environmental effect, EPA must develop standards to address these remaining risks. This second stage of the regulatory process is known as the residual risk stage. For each source category for which EPA issued MACT standards, the residual risk stage must be completed within eight years of promulgation of the initial technology-based standard. In December of 2006 we consulted with a panel from the EPA's Science Advisory Board (SAB) on the Risk and Technology Review (RTR) Assessment Plan and in June of 2007, we received a letter with the results of that consultation. Subsequent to the consultation, in July of 2009 a meeting was held with an SAB panel for a formal peer review of the Risk and Technology Review (RTR) Assessment Methodologies [2]. We received the final SAB report on this review in May of 2010 [3]. Where appropriate, we have responded to the key messages from this review in developing our current risk assessments and we will be continuing our efforts to improve our assessments by incorporating updates based on the SAB recommendations as they are developed and become available. Our responses to the key recommendations of the SAB are outlined in the memo for this rulemaking docket [4]. This document contains the methods and the results of baseline risk assessments (i.e., after the implementation of the respective MACT standards) performed for the oil and natural gas production and natural gas transmission and storage source categories. The methods discussion includes descriptions of the methods used to develop refined estimates of chronic inhalation exposures and human health risks for cancer and noncancer endpoints, as well as descriptions of the methods used to screen for acute health risks, chronic non-inhalation health risks, and adverse environmental effects. Since the screening assessments indicated low potential for chronic non-inhalation health effects or environmental impacts, including effects to threatened and endangered species, no further refinement of these assessments was performed.


5

2 Methods2.1 Emissions and source dataThe 2005 National-Scale Air Toxics Assessment (NATA) National Emissions Inventory (NEI) served as the starting point for this assessment. The 2005 NEI contains information on actual emissions during the entire 2005 base year. Using the process MACT codes1, we developed a subset of this inventory that contains emissions and facility data for the oil and natural gas production and natural gas transmission and storage source categories. Next, we performed an engineering review of the information using EPA engineers who were directly involved in the development of the MACT standards for the source category, and/or who have extensive knowledge of the characteristics of this industry. The NEI was updated with industry supplied data as available. The goal of the engineering review was to identify readily-apparent limitations and issues with the emissions data (particularly those that would have the potential to influence risk estimates) and to make changes to the data set where possible to address these issues and decrease the uncertainties associated with the assessment. Details on the development of the emissions and source data for this source category are discussed in Section 3. The emissions data and modifications made to the NEI data are discussed in Appendix 1, entitled Emissions Inventory Support Memorandum.

2.2 Dispersion modeling for inhalation exposure assessmentBoth long- and short-term inhalation exposure concentrations and associated health risk from each facility in the source category of interest were estimated using the Human Exposure Model in combination with the American Meteorological Society/EPA Regulatory Model dispersion modeling system (HEM-AERMOD). The approach used in applying this modeling system is outlined below, and further details are provided in Appendix 2. The HEMAERMOD performs three main operations: atmospheric dispersion modeling, estimation of individual human exposures and health risks, and estimation of population risks. This section focuses on the dispersion modeling component. The exposure and risk characterization components are discussed in other subsections of Sections 2 and 3. The dispersion model in the HEM-AERMOD system, AERMOD version 09292, is a state-ofthe-science Gaussian plume dispersion model that is preferred by EPA for modeling point, area, and volume sources of continuous air emissions from facility applications [5]. Further details on AERMOD can be found in the AERMOD Users Guide [6]. The model is used to develop annual average ambient concentrations through the simulation of hour-by-hour dispersion from the emission sources into the surrounding atmosphere. Hourly emission rates used for this simulation are generated by evenly dividing the total annual emission rate from the inventory into the 8,760 hours of the year.

The tagging of data with MACT codes allows EPA to determine reductions attributable to the MACT program. The NEI associates MACT codes corresponding to MACT source categories with stationary major and area source data. MACT codes are assigned at the process level for the point source.

1

6 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE The first step in the application of the HEM-AERMOD modeling system is to predict ambient concentrations at locations of interest. The AERMOD model options employed are summarized in Table 2.2-1 and are discussed further below. Table 2.2-1 AERMOD version 09292 model options for RTR modelingModeling Option Type of calculations Source type Receptor orientation Terrain characterization Building downwash Plume deposition/depletion Urban source option Meteorology Selected Parameter for chronic exposure Hourly Ambient Concentration Point, area represented as pseudo point source Polar (13 rings and 16 radials) Discrete (census block centroids) Actual from USGS 1-degree DEM data Not Included Not Included No 1 year representative NWS from nearest site (over 200 stations)

In HEM-AERMOD, meteorological data are ordinarily selected from a list of over 200 National Weather Service (NWS) surface observation stations across the continental United States, Alaska, Hawaii, and Puerto Rico. In most cases the nearest station is selected as representative of the conditions at the subject facility. Ideally, when considering off-site meteorological data most site-specific dispersion modeling efforts will employ up to five years of data to capture variability in weather patterns from year to year. However, because we had an insufficient number of appropriately formatted model input files derived from available meteorological data, we modeled only a single year, typically 1991. While the selection of a single year may result in under-prediction of long-term ambient levels at some locations, likewise it may result in over-prediction at others. For each facility identified by its characteristic latitude and longitude coordinates, the closest meteorological station was used in the dispersion modeling. The average distance between a modeled facility and the applicable meteorological station was 40 miles (72 km). Appendix 3 (Meteorological Data Processing Using AERMET for HEM-AERMOD) provides a complete listing of stations and assumptions along with further details used in processing the data through AERMET. The sensitivity of model results to the selection of the nearest weather station and the use of one year of meteorological data is discussed in Risk and Technology Review (RTR) Risk Assessment Methodologies [2]. The HEM-AERMOD system estimates ambient concentrations at the geographic centroids of census blocks (using the 2000 Census), and at other receptor locations that can be specified by the user. The model accounts for the effects of multiple facilities when estimating concentration impacts at each block centroid. Typically we combined only the impacts of facilities within the same source category, and assessed chronic exposure and risk only for

7 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE census blocks with at least one resident (i.e., locations where people may reasonably be assumed to reside rather than receptor points at the fenceline of a facility). Chronic ambient concentrations were calculated as the annual average of all estimated short-term (one-hour) concentrations at each block centroid. Possible future residential use of currently uninhabited areas was not considered. Census blocks, the finest resolution available in the census data, are typically comprised of approximately 40 people or about ten households. In contrast to the development of ambient concentrations for evaluating long-term exposures, which was performed only for occupied census blocks, worst-case short-term (one-hour) concentrations were estimated both at the census block centroids and at points nearer the facility that represent locations where people may be present for short periods, but generally no nearer than 100 meters from the center of the facility (note that for large facilities, this 100-meter ring could still contain locations inside the facility property). Since short-term emission rates were needed to screen for the potential for hazard via acute exposures, and since the NEI contains only annual emission totals, we generally apply the assumption to all source categories that the maximum one-hour emission rate from any source is ten times the average annual hourly emission rate for that source. The average hourly emissions rate is defined as the total emissions for a year divided by the total number of operating hours in the year. The choice of a factor of ten for acute screening was originally based on engineering judgment. To develop a more robust peak-to-mean emissions factor, and in response to one of the key messages from the SAB consultation on our RTR Assessment Plan, we performed an analysis using a short-term emissions dataset from a number of sources located in Texas (originally reported on by Allen et al. 2004)[7]. In that report, the Texas Environmental Research Consortium Project compared hourly and annual emissions data for volatile organic compounds for all facilities in a heavilyindustrialized 4-county area (Harris, Galveston, Chambers, and Brazoria Counties, TX) over an eleven-month time period in 2001. We obtained the dataset and performed our own analysis, focusing that analysis on sources which reported emitting high quantities of HAP over short periods of time (see Appendix 4, Analysis of data on short-term emission rates relative to long-term emission rates). Most peak emission events were less than twice the annual average, the highest was a factor of 74 times the annual average, and the 99th percentile ratio of peak hourly emission rate to the annual hourly emission rate was 9. Based on these results, we chose the factor of ten for all initial screening; it is intended to cover routinely-variable emissions as well as startup, shutdown, and malfunction (SSM) emissions. While there have been some documented emission excursions above this level, our analysis of the data from the Texas Environmental Research Consortium suggests that this factor should cover more than 99% of the short-term peak gaseous or volatile HAP emissions from typical industrial sources. Census block elevations for HEM-AERMOD modeling were determined nationally from the US Geological Service 1-degree digital elevation model (DEM) data files, which have a spatial resolution of about 90 meters. Elevations of polar grid points used in estimating shortand long-term ambient concentrations were assumed to be equal to the highest elevation of any census block falling within the polar grid sector corresponding to the grid point. If a sector does not contain any blocks, the model defaults the elevation to that of the nearest

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE block. If an elevation is not provided for the emission source, the model uses the average elevation of all sectors within the innermost model ring.

8

In addition to using receptor elevation to determine plume height, AERMOD adjusts the plumes flow if nearby elevated hills are expected to influence the wind patterns. For details on how hill heights were estimated and used in the AERMOD modeling, see Appendix 2.

2.3 Estimating human inhalation exposureWe used the annual average ambient air concentration of each HAP at each census block centroid as a surrogate for the lifetime inhalation exposure concentration of all the people who reside in the census block. That is, the risk analysis did not consider either the short-term or long-term behavior (mobility) of the exposed populations and its potential influence on their exposure. We did not address short-term human activity for two reasons. First, our experience with the NATA assessments (which modeled daily activity using EPAs HAPEM model) suggests that, given our current understanding of microenvironment concentrations and daily activities, modeling short-term activity would, on average, reduce risk estimates about 25% for particulate HAPs; it will also reduce risk estimates for gaseous HAPs, but typically by much less. Second, basing exposure estimates on average ambient concentrations at census block centroids may underestimate or overestimate actual exposure concentrations at some residences. Further reducing exposure estimates for the most highly exposed residents by modeling their short-term behavior could add a systematic low bias to these results. We did not address long-term migration nor population growth or decrease over 70 years, instead basing the assessment on the assumption that each persons predicted exposure is constant over the course of their lifetime which is assumed to be 70 years. In assessing cancer risk, we generally estimated three metrics; the maximum individual risk (MIR), which is defined as the risk associated with a lifetime of exposure at the highest concentration; the population risk distribution; and the cancer incidence. The assumption of not considering short or long-term population mobility does not bias the estimate of the theoretical MIR nor does it affect the estimate of cancer incidence since the total population number remains the same. It does, however, affect the shape of the distribution of individual risks across the affected population, shifting it toward higher estimated individual risks at the upper end and reducing the number of people estimated to be at lower risks, thereby increasing the estimated number of people at specific risk levels. When screening for potentially significant acute exposures, we used an estimate of the highest hourly ambient concentration at any off-site location as the surrogate for the maximum potential acute exposure concentration for any individual.

2.4 Acute Risk Screening and Refined AssessmentsIn establishing a scientifically defensible approach for the assessment of potential health risks due to acute exposures to HAP, we followed the same general approach that has been used for developing chronic health risk assessments under the residual risk program. That is, we developed a tiered, iterative approach. This approach to risk assessment was endorsed by the

9 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE National Academy of Sciences in its 1993 publication Science and Judgment in Risk Assessment and subsequently was adopted in the EPAs Residual Risk Report to Congress in 1999. The assessment methodology is designed to eliminate from further consideration those facilities for which we have confidence that no acute adverse health effects of concern will occur. To do so, we use what is called a tiered, iterative approach to the assessment. This means that we begin with a screening assessment, which relies on readily available data and uses conservative assumptions that in combination approximate a worst-case exposure. The result of this screening process is that either the facility being assessed poses no acute health risks (i.e., it screens out), or that it requires further, more refined assessment. A refined assessment could use industry- or site-specific data on the temporal pattern of emissions, the layout of emission points at the facility, the boundaries of the facility, and/or the local meteorology. In some cases, all of these site-specific data would be needed to refine the assessment; in others, lesser amounts of site-specific data could be used to determine that acute exposures are not a concern, and significant additional data collection would not be necessary. Acute health risk screening was performed as the first step. We used conservative assumptions for emission rates, meteorology, and exposure location. We used the following worst-case assumptions in our screening approach: Peak 1-hour emissions were assumed to equal 10 times the average 1-hour emission rates. For facilities with multiple emission points, peak 1-hour emissions were assumed to occur at all emission points at the same time. For facilities with multiple emission points, 1-hour concentrations at each receptor were assumed to be the sum of the maximum concentrations due to each emission point, regardless of whether those maximum concentrations occurred during the same hour. Worst-case meteorology (from one year of local meteorology) was assumed to occur at the same time the peak emission rates occur. The recommended EPA local-scale dispersion model, AERMOD, is used for simulating atmospheric dispersion. A person was assumed to be located downwind at the point of maximum impact during this same 1-hour period, but no nearer to the source than 100 meters. The maximum impact was compared to multiple short-term health benchmarks for the chemical being assessed to determine if a possible acute health risk might exist. These benchmarks are described in section 2.6 of this report.

As mentioned above, when we identify acute impacts which exceed their relevant benchmarks, we pursue refining our acute screening estimates. In some cases, this includes use of a refined emissions multiplier to estimate the peak hourly emission rates from the average rates. For the oil and gas production and natural gas transmission and storage source categories, we conducted a review of the layout of emission points at the facilities with the facility boundaries to determine the maximum off-site acute impact for the facilities that did

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE not screen out during the initial model run. Refer to Appendices 6 and 7 for the detailed results for these sites.

10

2.5 Multipathway and environmental risk screeningThe potential for significant human health risks due to exposures via routes other than inhalation (i.e., multipathway exposures) was screened by first determining whether any sources emitted any hazardous air pollutants known to be persistent and bioaccumulative in the environment (PB-HAP). The PB-HAP compounds or compound classes are identified for the screening from the EPAs Air Toxics Risk Assessment Library [8]. Examples of PB-HAP are cadmium compounds, chlordane, chlorinated dibenzodioxins and furans, DDE, heptachlor, hexachlorobenzene, hexachlorocyclohexane, lead compounds, mercury compounds, methoxychlor, polychlorinated biphenyls, polycyclic organic matter (POM), toxaphene, and trifluralin. Emissions of POM were identified in the emissions inventories for the natural gas transmission and storage source category and also for the oil and natural gas production source category. These emissions were evaluated for potential non-inhalation risks and adverse environmental impacts using our recently-developed screening scenario which was developed for use with the TRIM.FaTE2 model. This screening scenario uses environmental media outputs from the peer-reviewed TRIM.FaTE to estimate the maximum potential ingestion risks for any specified emission scenario by using a generic farming/fishing exposure scenario that simulates a subsistence environment. The screening scenario retains many of the ingestion and scenario inputs developed for EPAs Human Health Risk Assessment Protocols (HHRAP) for hazardous waste combustion facilities.3 In the development of the screening scenario a sensitivity analysis was conducted to ensure that its key design parameters were established such that environmental media concentrations were not underestimated, and to also minimize the occurrence of false positives for human health endpoints. See Appendix 5 for a complete discussion of the development and testing of the screening scenario, as well as for the values of facility-level de minimis emission rates developed for screening potentially significant multi-pathway impacts. For the purpose of developing de minimis emission rates for our multi-pathway screening, we derived emission levels for each PB-HAP at which the maximum human health risk would be 1 in a million for lifetime cancer risk or a hazard quotient of 1.0 for noncancer impacts. In evaluating the potential multi-pathway risks from emissions of lead compounds, rather than developing a de minimis emission rate, we compared maximum estimated chronic (3-month average) atmospheric concentrations with the current National Ambient Air Quality Standard (NAAQS) for lead. Values below the NAAQS were considered to have a low potential for multi-pathway risks of any significance. Lead was not reported as being emitted from the source categories assessed in this risk document. There was only one facility in the natural gas transmission and storage source category with reported emissions of PB-HAP, and the emission rates were less than the de minimis emission rates. There were 29 facilities in the oil and gas production source category with reported emissions of PB-HAP, and one of these had emission rates greater than the de minimis2 3

EPAs Total Risk Integrated Methodology (General Information) http://epa.gov/ttn/fera/trim_gen.html EPAs Human Health Risk Assessment Protocol (HHRAP) for Hazardous Waste Combustion Facilities; http://www.epa.gov/epaoswer/hazwaste/combust/riskvol.htm#volume1

11 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE emission rates. In this case, the de minimis emission rate for POM was exceeded by a factor of six. For POM, dairy, vegetables, and fruits were the three most dominant exposure pathways driving human exposures in the hypothetical screening exposure scenario. The single facility with emissions exceeding the de minimis emission rate for POM is located in a highly industrialized area. Therefore, the exposure pathways driving human exposure are unlikely. For the reasons discussed above, multi-pathway exposures and environmental risks were deemed negligible and no further analysis was performed. Additionally, we evaluated the potential for significant ecological exposures to non PB-HAP from exceedances of chronic human health inhalation thresholds in the ambient air near these facilities. Human health dose-response threshold values are generally derived from studies conducted on laboratory animals (such as rodents) and developed with the inclusions of uncertainty factors that could be as high as 3000. As a result, these human threshold values are often significantly lower than the level expected to cause an adverse effect in an exposed rodent. It should be noted that there is a scarcity of data on the direct atmospheric impact of these HAPs on other receptors, such as plants, birds, and wildlife. Thus, if the maximum inhalation hazard in an ecosystem is below the level of concern for humans, we have generally concluded that mammalian receptors should be at no risk of adverse effects due to inhalation exposures from non PB-HAP, and have assurance that other ecological receptors are also not at any significant risk from direct atmospheric impact. In some isolated cases where we have data indicating potential adverse impacts on plants, birds, or other wildlife due to the direct atmospheric impacts of specific HAPs, we note that as an uncertainty and, where possible, refine our analysis by comparing our modeled impacts to available threshold values from the scientific literature.

2.6 Dose-Response Assessment2.6.1 Sources of chronic dose-response information Dose-response assessment (carcinogenic and non-carcinogenic) for chronic exposure (either by inhalation or ingestion) for the HAPs reported in the emissions inventory for the oil and gas production and the natural gas transmission and storage source categories were based on the EPA Office of Air Quality Planning and Standards existing recommendations for HAPs [9], also used for NATA [10]. This information has been obtained from various sources and prioritized according to (1) conceptual consistency with EPA risk assessment guidelines and (2) level of peer review received. The prioritization process was aimed at incorporating into our assessments the best available science with respect to dose-response information. The recommendations are based on the following sources, in order of priority: 1) US Environmental Protection Agency (EPA). EPA has developed dose-response assessments for chronic exposure for many of the pollutants in this study. These assessments typically provide a qualitative statement regarding the strength of scientific data and specify a reference concentration (RfC, for inhalation) or reference dose (RfD, for ingestion) to protect against effects other than cancer and/or a unit risk estimate (URE, for inhalation) or slope factor (SF, for ingestion) to estimate the probability of developing cancer. The RfC is defined as an estimate (with uncertainty spanning perhaps an order of

12 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. The RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. The URE is defined as the upper-bound excess cancer risk estimated to result from continuous lifetime exposure to an agent at a concentration of 1 g/m3 in air. The SF is an upper bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime exposure to an agent. This estimate, [is] usually expressed in units of proportion (of a population) affected per mg/kg-day EPA disseminates dose-response assessment information in several forms, based on the level of review. The Integrated Risk Information System (IRIS) [11] is an EPA database that contains scientific health assessment information, including dose-response information. All IRIS assessments since 1996 have also undergone independent external peer review. The current IRIS process includes review by EPA scientists, interagency reviewers from other federal agencies, and the public, and peer review by independent scientists external to EPA. EPAs science policy approach, under the current carcinogen guidelines, is to use linear low-dose extrapolation as a default option for carcinogens for which the mode of action (MOA) has not been identified. We expect future EPA dose-response assessments to identify nonlinear MOAs where appropriate, and we will use those analyses (once they are peer reviewed) in our risk assessments. At this time, however, there are no available carcinogen dose-response assessments for inhalation exposure that are based on a nonlinear MOA. 2) US Agency for Toxic Substances and Disease Registry (ATSDR). ATSDR, which is part of the US Department of Health and Human Services, develops and publishes Minimum Risk Levels (MRLs) [12] for inhalation and oral exposure to many toxic substances. As stated on the ATSDR web site: Following discussions with scientists within the Department of Health and Human Services (HHS) and the EPA, ATSDR chose to adopt a practice similar to that of the EPA's Reference Dose (RfD) and Reference Concentration (RfC) for deriving substance specific health guidance levels for non neoplastic endpoints. The MRL is defined as an estimate of daily human exposure to a substance that is likely to be without an appreciable risk of adverse effects (other than cancer) over a specified duration of exposure. ATSDR describes MRLs as substance-specific estimates to be used by health assessors to select environmental contaminants for further evaluation. Exposures above an MRL do not necessarily represent a threat, and MRLs are therefore not intended for use as predictors of adverse health effects or for setting cleanup levels. 3) California Environmental Protection Agency (CalEPA). The CalEPA Office of Environmental Health Hazard Assessment has developed dose-response assessments for many substances, based both on carcinogenicity and health effects other than cancer. The process for developing these assessments is similar to that used by EPA to develop IRIS values and incorporates significant external scientific peer review. As cited in the CalEPA Technical Support Document for developing their chronic assessments4: The4

Air Toxics Hot Spots Program, Risk Assessment Guidelines, Part III - Technical Support Document

13 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE guidelines for developing chronic inhalation exposure levels incorporate many recommendations of the U.S. EPA [13] and NAS [14]. The non-cancer information includes available inhalation health risk guidance values expressed as chronic inhalation reference exposure levels (RELs) [15]. CalEPA defines the REL as the concentration level at or below which no health effects are anticipated in the general human population. CalEPA's quantitative dose-response information on carcinogenicity by inhalation exposure is expressed in terms of the URE [16], defined similarly to EPA's URE. In developing chronic risk estimates, we adjusted dose-response values for some HAPs based on professional judgment, as follows: 1) In the case of HAP categories such as glycol ethers and cyanide compounds, the most conservative dose-response value of the chemical category is used as a surrogate for other compounds in the group for which dose-response values are not available. This is done in order to examine, under conservative assumptions, whether these HAPs that lack doseresponse values may pose an unacceptable risk and require further examination, or screen out from further assessment. 2) Where possible for emissions of unspecified mixtures of HAP categories such as metal compounds and POM, we apply category-specific chemical speciation profiles appropriate to the source category to develop a composite dose-response value for the category. 3) In 2004, the EPA determined that the Chemical Industry Institute of Toxicology (CIIT) cancer dose-response value for formaldehyde (5.5 x 10-9 per g/m3) was based on better science than the IRIS cancer dose-response value (1.3 x 10-5 per g/m3), and we switched from using the IRIS value to the CIIT value in risk assessments supporting regulatory actions. However, subsequent research published by the EPA suggests that the CIIT model was not appropriate and in 2010 EPA returned to using the 1991 IRIS value, which is more health protective.[17 ] EPA has been working on revising the formaldehyde IRIS assessment and the National Academy of Sciences (NAS) completed its review of the EPAs draft in May of 2011. EPA is reviewing the public comments and the NAS independent scientific peer review, and the draft IRIS assessment will be revised and the final assessment will be posted on the IRIS database. In the interim, we will present findings using the 1991 IRIS value as a primary estimate, and may also consider other information as the science evolves. 4) A substantial proportion of POM reported to EPAs national emission inventory (NEI) are not speciated into individual compounds. As a result, it is necessary to apply the same simplifying assumptions to assessments that are used in NATA [18]. The NATA approach partitions POM into eight different non-overlapping groups that are modeled as separate pollutants. Each POM group comprises POM species of similar carcinogenic potency, for which we can apply the same URE.for the Determination of Non-cancer Chronic Reference Exposure Levels. Air Toxicology and Epidemiology Section, Office of Environmental Health Hazard Assessment, California Environmental Protection Agency. February 2000 (http://www.oehha.ca.gov/air/chronic_rels/pdf/relsP32k.pdf)

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE The emissions inventories for the oil and gas production and for the natural gas transmission and storage source categories include emissions of 58 HAP with available chronic quantitative inhalation dose-response values. These HAP, their doseresponse values, and the source of the values are listed in Tables 2.6-1 (a) and (b). Table 2.6-1 (a) Dose-Response Values for Chronic Inhalation Exposure to Carcinogens URE (unit risk estimate for cancer)5 = cancer risk per g/m3 of average lifetime exposure. Sources: IRIS = EPA Integrated Risk Information System, CAL = California EPA Office of Environmental Health Hazard Assessment. Pollutant Acetaldehyde Acrylamide Arsenic compounds Benzene7 Beryllium compounds p-Dichlorobenzene 1,4-Dioxane Ethyl benzene Ethylene dibromide Ethylene dichloride Ethylene oxide Formaldehyde Methylene chloride Methyl tert-butyl ether Naphthalene Propylene oxide Polycyclic Organic Matter - 2-Methylnaphthalene - 3-Methylcholanthrene - 7,12Dimethylbenz[a]Anthracene - Acenaphthene - Anthracene - Benz[a]Anthracene5

14

CAS Number6 75070 79061 7440382 71432 7440417 106467 123911 100414 106934 107062 75218 50000 75092 1634044 91203 75569 91576 56495 57976 83329 56553

URE5 (1/g/m3) 2.2E-06 1.0E-04 4.3E-03 7.8E-06 2.4E-03 1.1E-05 7.7E-06 2.5E-06 6.0E-04 2.6E-05 8.8E-05 1.3E-05 4.7E-07 2.6E-07 3.4E-05 3.7E-06 8.8E-05 6.3E-03 7.1E-02 8.8E-05 8.8E-05 1.1E-04

Source IRIS IRIS IRIS IRIS IRIS CAL CAL CAL IRIS IRIS CAL IRIS IRIS CAL CAL IRIS CAL CAL CAL CAL CAL CAL

The URE is the upper-bound excess cancer risk estimated to result from continuous lifetime exposure to an agent at a concentration of 1 g/m3 in air. UREs are considered upper bound estimates meaning they represent a plausible upper limit to the true value. 6 Chemical Abstract Services identification number. For groups of compounds that lack a CAS number we have used a surrogate 3-digit identifier corresponding to the groups position on the CAA list of HAPs. 7 The EPA IRIS assessment for benzene provides a range of equally plausible UREs. This assessment used the highest value in that range, 7.8E-06 per ug/m3. The low end of the range is 2.2E-06 per ug/m3.

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 2.6-1 (a) Dose-Response Values for Chronic Inhalation Exposure to Carcinogens URE (unit risk estimate for cancer)5 = cancer risk per g/m3 of average lifetime exposure. Sources: IRIS = EPA Integrated Risk Information System, CAL = California EPA Office of Environmental Health Hazard Assessment. Pollutant - Benzo[a]Pyrene - Benzo[b]Fluoranthene - Benzo[g,h,i]Perylene - Benzo[k]Fluoranthene - Chrysene - Dibenzo[a,h]Anthracene - Fluoranthene - Fluorene - Indeno[1,2,3-c,d]Pyrene - Phenanthrene - Pyrene CAS Number6 50328 205992 191242 207089 218019 53703 206440 86737 193395 85018 129000 URE5 (1/g/m3) 1.1E-03 1.1E-04 8.8E-05 1.1E-04 1.1E-05 1.2E-03 8.8E-05 8.8E-05 1.1E-04 8.8E-05 8.8E-05 Source CAL CAL CAL CAL CAL CAL CAL CAL CAL CAL CAL

15

Table 2.6-1 (b) Dose-Response Values for Chronic Inhalation Exposure to Noncarcinogens RfC (reference inhalation concentration) = an estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. Sources: IRIS = EPA Integrated Risk Information System, CAL = California EPA Office of Environmental Human Health Assessment, ATSDR = US Agency for Toxic Substances Disease Registry, Pollutant CAS Number6 RfC Source8 (mg/m3) Acetaldehyde 75070 0.009 IRIS -- L Acrolein 107028 0.00002 IRIS -- H Acrylamide 79061 0.006 IRIS -- M Arsenic compounds 7440382 0.000015 CAL Benzene 71432 0.03 IRIS -- M Beryllium compounds 7440417 0.00002 IRIS -- M Carbon disulfide 75150 0.7 IRIS -- M Chlorobenzene 108907 1 CAL Chloroform 67663 0.098 ATSDRThe descriptors L (low), M (medium), and H (high) have been added for IRIS RfC values to indicate the overall level of confidence in the RfC value, as reported in IRIS.8

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 2.6-1 (b) Dose-Response Values for Chronic Inhalation Exposure to Noncarcinogens

16

RfC (reference inhalation concentration) = an estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. Sources: IRIS = EPA Integrated Risk Information System, CAL = California EPA Office of Environmental Human Health Assessment, ATSDR = US Agency for Toxic Substances Disease Registry, Pollutant CAS Number6 RfC Source8 (mg/m3) Cresols (mixed) 1319773 0.6 CAL -o-Cresol 95487 0.6 CAL Cumene 98828 0.4 IRIS -- H/M p-Dichlorobenzene 106467 0.8 IRIS -- M Diethanolamine 111422 0.003 CAL 1,4-Dioxane 123911 3.6 D-ATSDR Ethylene dibromide 106934 0.009 IRIS -- M Ethyl benzene 100414 1 IRIS -- L Ethylene dichloride 107062 2.4 ATSDR Ethylene Glycol 107211 0.4 CAL Ethylene Oxide 75218 0.03 CAL Formaldehyde 50000 0.0098 ATSDR Glycol Ethers 9 - Ethylene glycol ethyl ether 110805 0.2 IRIS -- M - Ethylene glycol methyl ether 109864 0.02 IRIS -- M - Triethylene glycol 112276 0.02 IRIS -- M n-Hexane 110543 0.7 IRIS -- M Hydrochloric acid 7647010 0.02 IRIS -- L Methanol 67561 4 CAL Methyl bromide 74839 0.005 IRIS -- H Methylene chloride 75092 1 ATSDR Naphthalene 91203 0.003 IRIS -- M Phenol 108952 0.2 CAL Propylene oxide 75569 0.03 IRIS -- M Styrene 100425 1 IRIS -- M Toluene 108883 5 IRIS -- H Methyl Chloroform (1,1,1Trichloroethane) 71556 5 IRIS -- H/M Vinylidene chloride 75354 0.2 IRIS -- H/M Xylenes (mixed) 1330207 0.1 IRIS -- M m-Xylene10 108383 0.1 IRIS -- MThe RfC value for ethylene glycol methyl ether (EGME) was used as a surrogate for all glycol ethers without an RfC (denoted with an *).9

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 2.6-1 (b) Dose-Response Values for Chronic Inhalation Exposure to Noncarcinogens

17

RfC (reference inhalation concentration) = an estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. Sources: IRIS = EPA Integrated Risk Information System, CAL = California EPA Office of Environmental Human Health Assessment, ATSDR = US Agency for Toxic Substances Disease Registry, Pollutant CAS Number6 RfC Source8 (mg/m3) o-Xylene10 95476 0.1 IRIS -- M 10 p-Xylene 106423 0.1 IRIS -- M 2.6.2 Sources of acute dose-response information Hazard identification and dose-response assessment information for preliminary acute inhalation exposure assessments are based on the existing recommendations of OAQPS for HAPs [19]. Depending on availability, the results from screening acute assessments are compared to both no effects reference levels for the general public, such as the California Reference Exposure Levels (RELs), as well as emergency response levels, such as Acute Exposure Guideline Levels (AEGLs) and Emergency Response Planning Guidelines (ERPGs), with the recognition that the ultimate interpretation of any potential risks associated with an estimated exceedance of a particular reference level depends on the definition of that level and any limitations expressed therein. Comparisons among different available inhalation health effect reference values (both acute and chronic) for selected HAPs can be found in a newly released EPA document [20]. California Acute Reference Exposure Levels (RELs). The California Environmental Protection Agency (CalEPA) has developed acute dose-response reference values for many substances, expressing the results as acute inhalation Reference Exposure Levels (RELs). The acute REL (http://www.oehha.ca.gov/air/pdf/acuterel.pdf) is defined by CalEPA as the concentration level at or below which no adverse health effects are anticipated for a specified exposure duration. [21]. RELs are based on the most sensitive, relevant, adverse health effect reported in the medical and toxicological literature. RELs are designed to protect the most sensitive individuals in the population by the inclusion of margins of safety. Since margins of safety are incorporated to address data gaps and uncertainties, exceeding the REL does not automatically indicate an adverse health impact. Acute RELs are developed for 1-hour (and 8-hour) exposures. The values incorporate uncertainty factors similar to those used in deriving EPAs Inhalation Reference Concentrations (RfCs) for chronic exposures (and, in fact, California also has developed chronic RELs).

10

The RfC for mixed xylene was used as a surrogate.

18 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Acute Exposure Guideline Levels (AEGLs). AEGLs are developed by the National Advisory Committee (NAC) on Acute Exposure Guideline Levels (NAC/AEGL) for Hazardous Substances, and then reviewed and published by the National Research Council As described in the Committees Standing Operating Procedures (SOP) (http://www.epa.gov/opptintr/aegl/pubs/sop.pdf), AEGLs represent threshold exposure limits for the general public and are applicable to emergency exposures ranging from 10 min to 8 h. Their intended application is for conducting risk assessments to aid in the development of emergency preparedness and prevention plans, as well as real time emergency response actions, for accidental chemical releases at fixed facilities and from transport carriers. The document states that the primary purpose of the AEGL program and the NAC/AEGL Committee is to develop guideline levels for once-in-a-lifetime, short-term exposures to airborne concentrations of acutely toxic, high-priority chemicals. In detailing the intended application of AEGL values, the document states that, It is anticipated that the AEGL values will be used for regulatory and nonregulatory purposes by U.S. Federal and State agencies, and possibly the international community in conjunction with chemical emergency response, planning, and prevention programs. More specifically, the AEGL values will be used for conducting various risk assessments to aid in the development of emergency preparedness and prevention plans, as well as real-time emergency response actions, for accidental chemical releases at fixed facilities and from transport carriers. The NAC/AEGL defines AEGL-1 and AEGL-2 as: AEGL-1 is the airborne concentration (expressed as ppm or mg/m3) of a substance above which it is predicted that the general population, including susceptible individuals, could experience notable discomfort, irritation, or certain asymptomatic nonsensory effects. However, the effects are not disabling and are transient and reversible upon cessation of exposure. AEGL-2 is the airborne concentration (expressed as ppm or mg/m3) of a substance above which it is predicted that the general population, including susceptible individuals, could experience irreversible or other serious, long-lasting adverse health effects or an impaired ability to escape. Airborne concentrations below AEGL-1 represent exposure levels that can produce mild and progressively increasing but transient and nondisabling odor, taste, and sensory irritation or certain asymptomatic, nonsensory effects. With increasing airborne concentrations above each AEGL, there is a progressive increase in the likelihood of occurrence and the severity of effects described for each corresponding AEGL. Although the AEGL values represent threshold levels for the general public, including susceptible subpopulations, such as infants, children, the elderly, persons with asthma, and those with other illnesses, it is recognized that individuals, subject to unique or idiosyncratic responses, could experience the effects described at concentrations below the corresponding AEGL. Emergency Response Planning Guidelines (ERPGs). The American Industrial Hygiene Association (AIHA) has developed Emergency Response Planning Guidelines (ERPGs) [22]

19 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE for acute exposures at three different levels of severity. These guidelines represent concentrations for exposure of the general population (but not particularly sensitive persons) for up to 1 hour associated with effects expected to be mild or transient (ERPG-1), irreversible or serious (ERPG-2), and potentially life-threatening (ERPG-3). ERPG values (http://www.aiha.org/1documents/Committees/ERP-erpglevels.pdf) are described in their supporting documentation as follows: Emergency Response Planning Guidelines (ERPGs) were developed for emergency planning and are intended as health based guideline concentrations for single exposures to chemicals. These guidelines (i.e., the ERPG Documents and ERPG values) are intended for use as planning tools for assessing the adequacy of accident prevention and emergency response plans, including transportation emergency planning and for developing community emergency response plans. The emphasis is on ERPGs as planning values: When an actual chemical emergency occurs there is seldom time to measure airborne concentrations and then to take action. ERPG-1 and ERPG-2 values are defined by AIHA as follows: ERPG-1 is the maximum airborne concentration below which it is believed that nearly all individuals could be exposed for up to 1 hour without experiencing other than mild transient adverse health effects or without perceiving a clearly defined, objectionable odor. ERPG-2 is the maximum airborne concentration below which it is believed that nearly all individuals could be exposed for up to 1 hour without experiencing or developing irreversible or other serious health effects or symptoms which could impair an individual's ability to take protective action. The emissions inventories for the oil and gas production and for the natural gas transmission and storage source categories include emissions of 31 HAP with relevant and available quantitative acute dose-response threshold values. These HAPs, their acute threshold values, and the source of the value are listed below in Table 2.6-2.

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 2.6-2 Dose-Response Values for Acute Exposure CAS Number 75070 107028 7440382 71432 7440417 92524 75150 463581 108907 67663 98828 123911 100414 106934 107062 75218 50000 110805 109864 112276 110543 7647010 67561 74839 71556 75092 108952 75569 100425 108883 1330207 AEGL-1 (1-hr) (mg/m3) 81 0.069 170 40 46 250 61 140 130 1.1 AEGL-2 (1-hr) (mg/m3) 490 0.23 2600 61 500 140 690 310 1500 1200 4800 180 81 17 ERPG-1 (mg/m3) 81 0.069 170 40 310

20

Pollutant Acetaldehyde Acrolein Arsenic compounds Benzene Beryllium compounds Biphenyl Carbon disulfide Carbonyl sulfide Chlorobenzene Chloroform Cumene 1,4-Dioxane Ethyl benzene Ethylene dibromide Ethylene dichloride Ethylene oxide Formaldehyde Glycol ethers11 - Ethylene Glycol Ethyl Ether - Ethylene Glycol Methyl Ether - Triethylene Glycol Hexane Hydrochloric acid Methanol Methyl bromide Methyl chloroform (1,1,1-Trichloroethane) Methylene chloride Phenol Propylene oxide Styrene Toluene Xylenes (mixed)11

ERPG-2 (mg/m3) REL 490 0.47 0.23 0.0025 0.0002 2600 1.3 0.025 500 6.2 0.15

200 1.1

810 81 17

0.055 0.37 0.093 0.093

2.7 690 1300 690 58 170 85 750 560

12000 33 2700 820 3300 1900 89 690 550 4500 4000

2.7 690 1300 690 58 170 85 190

33 2700 820 3300 1900 89 690 550 1130

2.1 28 3.9 68 14 5.8 3.1 21 37 22

The acute REL for ethylene glycol methyl ether (EGME) was used as a surrogate for glycol ether compounds without an acute REL.

Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 2.6-2 Dose-Response Values for Acute Exposure CAS Number 108383 95476 106423 AEGL-1 (1-hr) (mg/m3) AEGL-2 (1-hr) (mg/m3) ERPG-1 (mg/m3) ERPG-2 (mg/m3)

21

Pollutant m-xylene12 o-xylene12 p-xylene12

REL 22 22 22

2.7 Risk Characterization2.7.1 General The final product of the risk assessment is the risk characterization, in which the information from the previous steps is integrated and an overall conclusion about risk is synthesized that is complete, informative, and useful for decision makers. In general, the nature of this risk characterization depends on the information available, the application of the risk information and the resources available. In all cases, major issues associated with determining the nature and extent of the risk are identified and discussed. Further, the EPA Administrators March 1995 Policy for Risk Characterization [23] specifies that a risk characterization be prepared in a manner that is clear, transparent, reasonable, and consistent with other risk characterizations of similar scope prepared across programs in the Agency. These principles of transparency and consistency have been reinforced by the Agencys Risk Characterization Handbook [24], in 2002 by the Agencys information quality guidelines [25], and in the OMB/OSTP September 2007 Memorandum on Updated Principles for Risk Analysis13, and are incorporated in these assessments. Estimates of health risk are presented in the context of uncertainties and limitations in the data and methodology. Through our tiered, iterative analytical approach, we have attempted to reduce both uncertainty and bias to the greatest degree possible in these assessments, within the limitations of available time and resources. We provide summaries of risk metrics (including maximum individual cancer risks and noncancer hazards, as well as cancer incidence estimates) along with a discussion of the major uncertainties associated with their derivation to provide decision makers with the fullest picture of the assessment and its limitations. For each carcinogenic HAP included in an assessment that has a potency estimate available, individual and population cancer risks were calculated by multiplying the corresponding lifetime average exposure estimate by the appropriate URE. This calculated cancer risk is defined as the upper-bound probability of developing cancer over a 70-yr period (i.e., the12 13

The REL for mixed xylenes was used as a surrogate. Memorandum for the Heads of Executive Departments and Agencies - Updated Principles for Risk Analysis (September 19, 2007), From Susan E. Dudley, Administrator, Office of Information and Regulatory Affairs, Office of Management and Budget; and Sharon L. Hays, Associate Director and Deputy Director for Science, Office of Science and Technology Policy (http://www.whitehouse.gov/omb/memoranda/fy2007/m07-24.pdf)

22 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE assumed human lifespan) at that exposure. Because UREs for most HAPs are upper-bound estimates, actual risks at a given exposure level may be lower than predicted, and could be zero. For EPAs list of carcinogenic HAPs that act by a mutagenic mode-of-action [26], we applied EPAs Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens [27]. This guidance has the effect of adjusting the URE by factors of 10 (for children aged 0-1), 3 (for children aged 2-15), or 1.6 (for 70 years of exposure beginning at birth), as needed in risk assessments. In this case, this has the effect of increasing the estimated life time risks for these pollutants by a factor of 1.6. In addition, although only a small fraction of the total POM emissions may be reported as individual compounds, EPA expresses carcinogenic potency for compounds in this group in terms of benzo[a]pyrene equivalence, based on evidence that carcinogenic POM have the same mutagenic mechanism of action as does benzo[a]pyrene. For this reason, EPA implementation policy [28] recommends applying the Supplemental Guidance to all carcinogenic PAHs for which risk estimates are based on relative potency. Accordingly, we applied the Supplemental Guidance to all unspeciated POM mixtures. Increased cancer incidence for the entire receptor population within the area of analysis was estimated by multiplying the estimated lifetime cancer risk for each census block by the number of people residing in that block, then summing the results for the entire modeled domain. This lifetime population incidence estimate was divided by 70 years to obtain an estimate of the number of cancer cases per year. In the case of benzene, the high end of the reported cancer URE range was used in our assessments to provide a conservative estimate of potential cancer risks. Use of the high end of the range provides risk estimates that are approximately 3.5 times higher than use of the equally-plausible low end value. When estimated benzeneassociated risks exceed 1 in a million, we also evaluate the impact of using the low end of the URE range on our risk results. Unlike linear dose-response assessments for cancer, noncancer health hazards generally are not expressed as a probability of an adverse occurrence. Instead, risk for noncancer effects is expressed by comparing an exposure to a reference level as a ratio. The hazard quotient (HQ) is the estimated exposure divided by a reference level (e.g., the RfC). For a given HAP, exposures at or below the reference level (HQ1) are not likely to cause adverse health effects. As exposures increase above the reference level (HQs increasingly greater than 1), the potential for adverse effects increases. For exposures predicted to be above the RfC, the risk characterization includes the degree of confidence ascribed to the RfC values for the compound(s) of concern (i.e., high, medium, or low confidence) and discusses the impact of this on possible health interpretations. The risk characterization for chronic effects other than cancer is expressed in terms of the HQ for inhalation, calculated for each HAP at each census block centroid. As discussed above, RfCs incorporate generally conservative uncertainty factors in the face of uncertain extrapolations, such that an HQ greater than one does not necessarily suggest the onset of

23 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE adverse effects. The HQ cannot be translated to a probability that adverse effects will occur, and is unlikely to be proportional to adverse health effect outcomes in a population. Screening for potentially significant acute inhalation exposures also followed the HQ approach. We divided the maximum estimated acute exposure by each available short-term threshold value to develop an array of HQ values relative to the various acute endpoints and thresholds. In general, when none of these HQ values are greater than one, there is no potential for acute risk. In those cases where HQ values above one are seen, additional information is used to determine if there is a potential for significant acute risks. 2.7.2 Mixtures Since most or all receptors in these assessments receive exposures to multiple pollutants rather than a single pollutant, we estimated the aggregate health risks associated with all the exposures from a particular source category combined. To combine risks across multiple carcinogens, our assessments use the mixtures guidelines [29,30] default assumption of additivity of effects, and combine risks by summing them using the independence formula in the mixtures guidelines. In assessing noncancer hazard from chronic exposures, in cases where different pollutants cause adverse health effects via completely different modes of action, it may be inappropriate to aggregate HQs. In consideration of these mode-of-action differences, the mixtures guidelines support aggregating effects of different substances in specific and limited ways. To conform to these guidelines, we aggregated non-cancer HQs of HAPs that act by similar toxic modes of action, or (where this information is absent) that affect the same target organ. This process creates, for each target organ, a target-organ-specific hazard index (TOSHI), defined as the sum of hazard quotients for individual HAPs that affect the same organ or organ system. All TOSHI calculations presented here were based exclusively on effects occurring at the critical dose (i.e., the lowest dose that produces adverse health effects). Although HQs associated with some pollutants have been aggregated into more than one TOSHI, this has been done only in cases where the critical dose affects more than one target organ. Because impacts on organs or systems that occur above the critical dose have not been included in the TOSHI calculations, some TOSHIs may have been underestimated. As with the HQ, the TOSHI should not be interpreted as a probability of adverse effects, or as strict delineation of safe and unsafe levels. Rather, the TOSHI is another measure of the potential for adverse health outcomes associated with pollutant exposure, and needs to be interpreted carefully by health scientists and risk managers. Because of the conservative nature of the acute inhalation screening and the variable nature of emissions and potential exposures, acute impacts were screened on an individual pollutant basis, not using the TOSHI approach. 2.7.3 Facility-wide Risks To help place the source category risks in context, we examined facility-wide risks using

24 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE 2005 NEI data and modeling as described in Section 2.2. . For the facilities in these source categories, we estimated the maximum inhalation cancer and chronic non-cancer risks associated with all HAP emissions sources at the facility, including emissions sources that are not part of the source categories but that are located within a contiguous area and are under common control. We analyzed risks due to the inhalation of HAP for the populations residing within 50 kilometers of each facility. The results of these facility-wide assessments are summarized below in the Risk Characterization section of this document. The complete results of the facility-wide assessments are provided in Table 2 of Appendix 6.

3 Risk Results for the Natural Gas Transmission and Storage Source Category 3.1 Source Category Description and ResultsThe natural gas transmission and storage source category comprises the pipelines, facilities, and equipment used to transport and store natural gas products (hydrocarbon liquids and gases). Specific equipment used in natural gas transmission and storage includes the mains, valves, meters, boosters, regulators, storage vessels, glycol dehydrators, compressors (and their driving units and appurtenances), and equipment used for transporting gas from a production plant, delivery point of purchased gas, gathering system, storage area, or other wholesale source of gas to one or more distribution areas. Glycol dehydration unit reboiler vents represent one HAP emission point at natural gas transmission and storage facilities. Other possible emission points include process vents, storage vessels with flash emissions, pipeline pigging and storage of pipeline pigging wastes, combustion sources, and equipment leaks. We currently estimate that there are 286 natural gas transmission and storage facilities operating in the U.S. The data set contains 321 facilities identified with a natural gas transmission and storage MACT code in the 2005 NATA NEI, January 2011 version. All 321 of these facilities are identified as major sources in the NEI. The emissions for the natural gas transmission and storage source category data set (of 321 facilities) are summarized in Table 3.1-1. The total HAP emissions for the source category are approximately 700 tons per year. Based on these data, the HAP emitted in the largest quantities are: toluene, hexane, benzene, xylenes (mixed), ethylene glycol, methanol, ethyl benzene, and 2,2,4-trimethylpentane. Emissions of these eight HAP make up 99 percent of the total emissions by mass. Persistent and bioaccumulative HAP (PB-HAP) 14 reported as emissions from these facilities include polycyclic organic matter.

14

Persistent and bioaccumulative HAP are defined in the EPAs Air Toxics Risk Assessment Library, Volume 1, EPa-453K-04-001A, as referenced in the ANPRM and provided on the EPAs Technology Transfer Network website for Fate, Exposure, and Risk Assessment at http://www.epa.gov/ttn/fera/risk_atra_vol1.html.

25 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 3.1-1 Summary of Emissions from the Natural Gas Transmission and Storage Source Category and Availability of Dose-Response ValuesNumber of Facilities Reporting HAP (286 facilities in data set) 309 311 310 301 259 11 291 267 259 3 3 260 1 259 43 2 2 1 41 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Prioritized Inhalation Dose-Response Value Identified by OAQPSb Unit Risk Estimate for Cancer? Reference Concentration for Noncancer? Health Benchmark Values for Acute Noncancer? PBHAP?

HAPa

Emissions (tpy)

Toluene Hexane Benzene Xylenes (mixed) Ethylene glycol Methanol Ethyl benzene 2,2,4-Trimethylpentane Carbonyl sulfide p-Xylene m-Xylene Naphthalene o-Xylene Carbon disulfide Formaldehyde Methyl tert-butyl ether Cumene Triethylene glycol Acetaldehyde p-Dichlorobenzene Polycyclic organic matter PAH, total 2-Methylnaphthalene Phenanthrene 7,12Dimethylbenz[a]Anthracene Pyrene Fluoranthene Fluorene Anthracene 3-Methylcholanthrene Acenaphthene Benz[a]Anthracene Benzo[b]Fluoranthene Benzo[k]Fluoranthene Chrysene Indeno[1,2,3-c,d]Pyrene Benzo[a]Pyrene

196 171 141 112 39 21 18 5 1 0.7 0.6 0.5 0.3 0.2 0.06 0.04 0.03 0.01 0.003 0.00002 0.000005 0.0000003 0.0000002 0.0000002 0.00000007 0.00000004 0.00000004 0.00000003 0.00000002 0.00000002 0.00000002 0.00000002 0.00000002 0.00000002 0.00000002 0.00000002

26 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 3.1-1 Summary of Emissions from the Natural Gas Transmission and Storage Source Category and Availability of Dose-Response ValuesNumber of Facilities Reporting HAP (286 facilities in data set) 1 1 1 1 Prioritized Inhalation Dose-Response Value Identified by OAQPSb Unit Risk Estimate for Cancer? Reference Concentration for Noncancer? Health Benchmark Values for Acute Noncancer? PBHAP?

HAPa

Emissions (tpy)

Benzo[g,h,i]Perylene Dibenzo[a,h]Anthracene Arsenic compounds Beryllium compoundsa

0.00000002 0.00000002 0.000003 0.0000002

Specific dose-response values for each chemical are identified on EPAs Technology Transfer Network website for air toxics at http://www.epa.gov/ttn/atw/toxsource/summary.html.b

Notes for how HAP were speciated for risk assessment: For most metals, emissions reported as the elemental metal are combined with metal compound emissions (e.g., cadmium emissions modeled as cadmium & compounds). For emissions of any chemicals or chemical groups classified as polycyclic organic matter (POM), emissions were grouped into POM subgroups as found on EPAs Technology Transfer Network website for the 2002 NationalScale Air Toxics Assessment at http://www.epa.gov/nata2002/methods.html#pom. (Approach to Modeling POM).

3.2 Risk CharacterizationThis section presents the results of the risk assessment for the natural gas transmission and storage source category. The basic chronic inhalation risk estimates presented here are the maximum individual lifetime cancer risk, the maximum chronic hazard index, and the cancer incidence. We also present results from our acute inhalation impact screening in the form of maximum hazard quotients, as well as the results of our preliminary screen for potential noninhalation risks from PB-HAP. Also presented are the HAP drivers, which are the HAP that collectively contribute 90 percent of the maximum cancer risk or maximum hazard index at the highest exposure location, as well as a summary of the results of our facility-wide assessments and our analysis of risks associated with the maximum allowed emissions under the current MACT standards. A detailed summary of the facility-specific risk assessment results is available in Appendix 6. Tables 3.2-1 and 3.2-2 summarize the chronic and acute inhalation risk results for the natural gas transmission and storage source category. The results indicate that maximum lifetime individual cancer risks could be as high as 90 in a million (30 in a million based on the lower end of the benzene URE range), with benzene as the major contributor to the risk. The total estimated cancer incidence from the source category is 0.001 excess cancer cases per year (0.0003 excess cancer cases per year based on the lower end of the benzene URE range), or one case in every 1000 years. Approximately 110 people are estimated to have cancer risks at or above 10 in a million, and approximately 2,500 people are estimated to have cancer risks at or above 1 in a million as a result of the emissions from 15 facilities (use of the lower end of

27 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE the benzene URE range would further reduce these population estimates). The maximum chronic non-cancer TOSHI value for the source category could be up to 0.4 from emissions of benzene, indicating no significant potential for chronic noncancer impacts. Worst-case acute hazard quotients (HQs) were calculated for every HAP shown in Table 3.11 that has an acute benchmark, and the highest acute HQ value of 9 (based on the benzene acute REL) is shown in Table 3.2-1. For cases where the screening acute HQ was greater than 1, we further refined the estimates by determining the highest HQ value that is outside facility boundaries. Table 3.2-2 provides more information on the acute risk estimates for HAP that had an acute HQ greater than 1 for any benchmark. The highest refined worst-case acute HQ value is 5 (based on the benzene acute REL). This estimated worst-case acute impact is significantly below the 1-hour AEGL-1 (and ERPG-1) value, corresponding to an acute HQAEGL-1 of 0.04. We conducted a screening-level evaluation of the potential human health risks associated with emissions of PB-HAP. Reported emissions of PB-HAP were compared to de minimis emission thresholds established by EPA for the purposes of the RTR risk assessments. 15 The PB-HAP emitted by facilities in this category include POM as benzo(a)pyrene toxicity equivalence (TEQ) (1 facility). All POM as benzo(a)pyrene TEQ emissions were below the de minimis threshold levels.

15

ICF International. TRIM-Based Multipathway Screening Scenario. Prepared for U.S. Environmental Protection Agency, Research Triangle Park, NC. October 2008.

28 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 3.2-1 Summary of Source Category Level Inhalation Risks for Natural Gas Transmission and Storage ResultFacilities in Source Category Number of Facilities Estimated to be in Source 286 Category Number of Facilities Identified in the NEI and 321 Modeled in Preliminary Risk Assessment Cancer Risks Maximum Individual Lifetime Cancer Risk (in 1 30-9016 million) Number of Facilities with Maximum Individual Lifetime Cancer Risk: Greater than or equal to 100 in 1 million 0 Greater than or equal to 10 in 1 million 3 Greater than or equal to 1 in 1 million 15 Chronic Noncancer Risks Maximum Hazard Index 0.4 Number of Facilities with Maximum Immunological Hazard Index: Greater than 1 0 Acute Noncancer Screening Results 9 Maximum Acute Hazard Quotient 0.07 0.01 Number of Facilities With Potential for Acute 15 Effects Refined Acute Noncancer Results 5 Maximum Acute Hazard Quotient 0.04 Number of Facilities With Potential for Acute 7 Effects Population Exposure Number of People Living Within 50 Kilometers 58,000,000 of Facilities Modeled Number of People Exposed to Cancer Risk: Greater than or equal to 100 in 1 million 0 Greater than or equal to 10 in 1 million 110 Greater than or equal to 1 in 1 million 2,500 Number of People Exposed to Noncancer Immunological Hazard Index: Greater than 1 0 Estimated Cancer Incidence (excess cancer cases 0.0003-0.0011616 per year) Contribution of HAP to Cancer Incidence: benzene 94%

HAP Driversn/a n/a benzene n/a benzene benzene, naphthalene, ethyl benzene benzene n/a benzene (REL) benzene (AEGL-1) toluene (ERPG-2) benzene benzene (REL) benzene (AEGL-1) benzene n/a n/a n/a n/a n/a n/a n/a

16

As previously mentioned, the EPA IRIS assessment for benzene provides a range of equally-plausible UREs (2.2E-06 to 7.8E-06 per ug/m3), giving rise to ranges for the estimates of cancer MIR and cancer incidence.

29 Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmission and Storage Source Categories -- FOR PUBLIC COMMENT, DO NOT CITE OR QUOTE Table 3.2-2 Summary of Refined Acute Results for Natural Gas Transmission and Storage Facilities

Refined Results MAXIMUM ACUTE HAZARD QUOTIENTS ACUTE DOSE-RESPONSE VALUES

HAP Benzene

Max. 1hr. Air Conc. (mg/m3) 6

Based on REL 5

Based on AEGL-1/ ERPG-1 0.04

Based on AEGL-2/ ERPG-2 0.002

REL (mg/m3) 1.3

AEGL-1 (1-hr) (mg/m3) 170

ERPG-1 (mg/m3) 170

AEGL-2 (1-hr) (mg/m3) 2600

ERPG-2 (mg/m3) 2600

Notes on Refined Process: 1) The screening was performed for all emitted HAP with available acute dose-response values. Only those pollutants whose screening HQs were greater than 1 for at least one acute threshold value are shown in the table. 2) HAP with available acute dose-response values which are not in the table do not carry any potential for posing acute health risks, based on an analysis of currently available emissions data. Notes on Acute Dose-Response Values: REL California EPA reference exposure level for no adverse effects. Most, but not all, RELs are for 1-hour exposures. AEGL Acute exposure guideline levels represent emergency exposure (1-hour) limits for the general public. AEGL-1 is the exposure level above which it is predicted that the general population, including susceptible individuals, could experience effects that are notable discomfort, but which are transient and reversible upon cessation of exposure. AEGL-2 is the exposure level above which it is predicted that the general population, including susceptible individuals, could experience irreversible or other serious, long-lasting adverse health effects of an impaired ability to escape. EPRG Emergency Removal Program guidelines represent emergency exposure (1-hour) limits for the general public. ERPG-1 is the maximum level below which it is believed that nearly all individuals could be exposed for up to 1 hour without experiencing other than mild, transient adverse health effects. ERPG-2 is the maximum exposure below which it is believed that nearly all individuals could be exposed for up to 1 hour without experiencing or developing irreversible or other serious health effects or symptoms which could impair an individuals ability to take protective action.

The results of the facility-wide assessments are summarized in Table 3.2-3. The results indicate that 74 facilities with natural gas transmission and storage operations have a facilitywide cancer MIR greater than or equal to 1 in a million. The maximum facility-wide cancer MIR is 200 in a million for 2 facilities, driven by formaldehyde emissions from reciprocating internal combustion engines17. The source category contributes one percent or less to the MIR in both cases. There is also one facility with a maximum facility-wide cancer MIR of 100 in a million, with 90 percent of the risk driven by the source category. Table 3.2-3 Source Category Contribution to Facility-Wide Cancer Risks Natural Gas Transmission & Storage Source Category MIR Contribution to Facility-Wide MIR > 90%17

Number of Facilities Binned by Facility-Wide MIR (in 1 million)

EPA Draft Risk Assessment for the Oil and Gas Production and Natural Gas Transmissions and Storage Source Categories, July 2011

Documents

natural gas transmission

storage source category

source category description

risk results

storage source categories

source category contribution

doseresponse assessment

source data